It is currently Sat Nov 18, 2017 10:08 pm



Post new topic Reply to topic
Author Message
ekstra   ara
 Post subject: Arc Blast vs Arc Flash
PostPosted: Mon Feb 27, 2012 9:52 pm 
Offline

Joined: Wed Apr 14, 2010 10:44 pm
Posts: 6
Location: Southport, NC
This is an issue I am currently struggling with. On most of my 4kv switchgear I have high arc calcs 100-161 cal/cm2. The calcs are based off a 910mm working distance, I can increase the length of the racking handle to reduce the calcs below 40 cal/cm2. (Yes I do have remote racking on order) Here is my struggle, just because I lower the flash hazard I have not removed the arc blast hazard. I have seen where a 4kv door was blown clean off the hinges and I doubt an extra 36" would have helped if an employee had been racking the breaker at the time. How are others handling this? I have never seen an arc blast calculation posted on equipment.


Top
 Profile Send private message  
Reply with quote  
 Post subject:
PostPosted: Tue Feb 28, 2012 4:39 am 
Offline
Plasma Level
User avatar

Joined: Tue Oct 26, 2010 9:08 am
Posts: 1799
Location: North Carolina
1. Right now NFPA 70E has three methods codified into it to address arc blast:
A. As with arc flash, arc blast does decrease with distance (it's a spherical wave front). It might be a bit of a total guess, but that's the reason that the information stops at 40 cal/cm^2...recognition that above that level, arc blast is the major consideration. So yes, lengthening handles does help.
B. Administrative controls are also codified. Yes, the EEWP rule is painful because it forces management to make a decision every time whether or not the work is really, truly necessary or just an easy way out. So the amount of energized work decreases. In many cases, folks have also gone to the point of doing the same with >40 cal/cm^2.
C. Aside from increasing distance, you can also decrease time. In our case with a lot of 23kV, 4.1 kV, and 2.3 kV gear with excessively high incident energy ratings, we modify the upstream breaker settings so that the instantaneous trip setting is less than the arcing current. Coordination is screwed up but the arc flash incident energy is reduced to a very reasonable level. Not sure that this truly addresses arc blast but that's another problem.

Aside from these rules, there has been a long standing rule in many organizations (ours included) that requires employees to stand to the side away from the door. I've seen some video tests put out there demonstrating what happens when the door comes flying off (looks like video from Kinectrics).

The future is looking pretty good, too. The joint IEEE/NFPA test work that is being done is looking at arc blast as well. So far from what I've been able to glean from their data, it appears that the pressure is pretty uniform almost irrespective of the voltage and current. It is very percussive and after the initial venting of all of the air, the arc blast pretty much subsides. I have not seen any published data but based on this it appears that once the arc power (some formula of voltage and current) exceeds a certain amount unless the trip time is extremely short, arc blast is a constant. That's encouraging relative to Lee's equation that doesn't put an upper limit on it (and which rapidly turns into a very bad situation). This addresses killing someone due to the percussive force alone. As to doors flying off...this is a construction strength issue which aside from arc resistant gear, there is no easy solution for this. If arc blasts turn out to be a constant, someone (NFPA or IEEE) can probably put out a performance test for "blast resistant gear" describing a simulated pressure wave and some performance requirements (doors stay on). The standard would be similar to arc resistant gear or surge suppressor tests with the 8x20 wave.


Top
 Profile Send private message  
Reply with quote  
 Post subject:
PostPosted: Thu Mar 01, 2012 2:34 am 
Offline

Joined: Wed Apr 14, 2010 10:44 pm
Posts: 6
Location: Southport, NC
I really appreciate all of that information. Now I have another. Seeming as it is a spherical wave front, would placing the operator farther away from the front of the cubical actually be placing him in the arc blast area? The operator's core would be out of the blast area if they were racking with a shorter handle and off to the side. Seems the arc flash would be spherical as well.
I am currently in a nuclear facility, so changing of settings and de-energizing busses to rack is not an easy feat (now that I think about it nothing is easy here).
Am I wrong in the thinking you would need to know what material and how much of it is available at the arc to determine the actual pressure that would be produced? Here is my thinking (for what its worth) assuming that all else is equal voltage, current, time... an arc blast produced from a 500 mcm cable would be less that an arc blast on a switchgear bus bar. Just for the simple fact that there is less material to vaporize. The expansion rate of material that vaporizes would seem to be important component of the equation as well.


Top
 Profile Send private message  
Reply with quote  
 Post subject:
PostPosted: Thu Mar 01, 2012 4:49 am 
Offline
Plasma Level
User avatar

Joined: Tue Oct 26, 2010 9:08 am
Posts: 1799
Location: North Carolina
Seeming as it is a spherical wave front, would placing the operator farther away from the front of the cubical actually be placing him in the arc blast area?

From the models or a practical point of view? Your general impression is correct. Close in, geometry matters a great deal. As you get further away however the differences in geometry tend to disappear. Mersen has some technical papers on their web site with some experiments documenting both the impact of adding phase barriers to a cubicle (the IEEE 1584 data set does not include these) as well as measuring incident energy "off axis".

However, the problem with Mersen's experiment is that it documented that this does actually occur in reality. But the data set is not large enough to develop a model to predict cases other than the simulated one. The old practice of "stand to one side" is very much validated by their data. Similarly, having doors open vs. doors closed does make a significant difference. The inherent difficulty is that we don't have enough model or engineering data to develop a standard based on this (except arc resistant gear for medium voltage which is a performance test).

You are wrong by the way about type and amount of material. Material type (copper vs. say carbon) does matter a little to arc physics because it changes the voltage drop at the electrodes. Since we're on the subject of arc flash though generally it is assumed that the electrodes are copper for good reason. Most circuit breaker design books talk about Hersha Ayrton's equation which has the voltage drop information. The amount does not matter because copper expands tens of thousands of times it's original volume when it changes from solid to gas. At that expansion rate, whether the wire is 14 gauge or 500 MCM doesn't matter. Other metals behave similarly.

I am at a chemical/mining facility and my uncle is an operator at the quad cities facility on the IL/IA border so I'm somewhat understanding of nuclear facilities...I've gotten a "behind the scenes" tour of a BWR in the past. You CAN change settings. You simply have to go through the entire safety analysis with a procedure in hand (make the procedure writers work for a change). If you lower settings or turn instantaneous tripping on, the risk is loss of coordination and the "what if's" due to lower settings (affecting large loads). The only major difference between PSM facilities and nuclear ones (other than the obvious...the process) is that PSM facilities can often use 10^-5 as a target failure rate while nuclear usually jacks it up to 10^-6 or even 10^-7. Failure to maintain breakers by the way if you go through the data in IEEE 493 increases the failure rate (and especially the personnel hazard) by about 10 fold. Draw out gear that is already marginal from a safety point of view ends up being unsafe under most safety codes. This happens because draw out gear is much more complex mechanically and thus subject to a lot more failures. The numbers I've seen in passing suggest that on drawout gear, the breaker itself fails 20% of the time. The racking mechanism fails 80% of the time. So just going to bolted gear decreases the failure rate 5 fold, without considering any other design differences.

Your own risk assessment procedures require you to analyze the setting changes by the way. Lowering settings on a breaker is an engineering and procedural solution. PPE is the last line of defense, to be used only if there are no other alternatives. All risk assessment procedures, nuclear or otherwise, follow this approach. Looking at PPE as the only solution is a shortcut, but a dangerous one. NFPA 70E Committee even states that for switching activities with certain gear, the risk of arc flash is remote. Applying data from a decent source (such as IEEE 493) seems to suggest that this analysis is correct even when applying probabilistic methods. Once you get into drawout gear though the inherent unreliability is really pushing what is considered acceptable.

Watch out for remote racking by the way. It sounds good but we've tried a couple of them. They tend to be poorly built or (because of a lack of protection against jams) can screw up a breaker. Even some OEM solutions have this problem.


Top
 Profile Send private message  
Reply with quote  
 Post subject:
PostPosted: Sat Mar 03, 2012 3:07 am 
Offline

Joined: Wed Apr 14, 2010 10:44 pm
Posts: 6
Location: Southport, NC
That whole reality thing can put a crimp in a good theory every time. Yes I know I can make changes in relay settings and I have all intentions to push for those changes. Making those changes is really difficult due to cordination that is required for nuclear safety. That being said, it is a long term plan. I am currently focusing on the low hanging fruit.

The remote racking I have selected is the Inolect system. I was impressed with its quality and I am not usually impressed easily. It monitors turns and current draw to avoid damage to the system. I will have them in service this July and will provide updates. My current plan is to implement remote racking and then surrond the area with an arc flash curtain. If anyone has any experience with this I'd love some input.

My thought about different metal types was along the lines of a conductor coming into contact with a steel support or similiar situation.

I am still struggling with the amount of copper not affecting the blast. My thinking was that the blast was caused by the vaporization of the copper. So it would seem to me that the more vaporization that happens the larger the blast would be ie. 1 in^3 of copper would equal ~50,000 in^3 of gas and 2 in^3 = ~100,000 in^3. Would the amount that vaporizes not affect the pressure created?

I really do appreciate your input. I am still pretty new to this side of electrial safety. Its nice to know there are people out there that will give advice to this degree.


Top
 Profile Send private message  
Reply with quote  
Display posts from previous:  Sort by  
Post new topic Reply to topic  [ 5 posts ] 

All times are UTC - 7 hours


Who is online

Users browsing this forum: No registered users and 1 guest


You cannot post new topics in this forum
You cannot reply to topics in this forum
You cannot edit your posts in this forum
You cannot delete your posts in this forum
You cannot post attachments in this forum

Jump to:  
cron
© 2017 Arcflash Forum / Brainfiller, Inc. | P.O. Box 12024 | Scottsdale, AZ 85267 USA | 800-874-8883